I guess it is better to find it out sooner rather than later: http://www.reddit.com/r/Bitcoin/comments/1v6law/ghash_41_again/ People are freaking out because one of the mining pools are reaching a relative high % of the mining hasing power. Currently it is around 41% and once it reaches 51%, the blockchain can be forked and coins can be double spent. Since that particular pool has no mining fees, people keep joining further increasing the problem. In plain English, right now the whole existence of bitcoin is depending on the average miners' logic to see that what might be good for me in the short run can hurt everyone in the long run. In even plainer English, the protocol has serious flaws and it should be fixed...
hehe good ol pekelo, dont be so dramatic ghasho.io is known for a long time and not a big deal, they are the farm running the http://cex.io miner shares (pretty nifty idea but way overpriced). The pool has vested interest to keep bitcoin alive, why would they try to destroy it. It's also not an instant kill switch once it hits 51%, the miners will know very quickly if the pool is doing something shady and they can just pull out, everything is out in the open. You will have a brief moment where the block transactions cannot be trusted but the history and everything will still be intact, once the pool shrinks all will be back to normal. The real risk of bitcoin is: 1) US/Europe government bans - imagine if bitcoin really took off and start to threaten the dollar dominance, imagine if commodities are looking to trade against bitcoins instead of dollars...you bet us government will shut it down, we have gone to wars for much less. 2) Banks or other private entities feel threatened if bitcoin becomes popular enough to jeopardize their fees (wire transfer, atm, credit card, you name it). At it is now, it will take < $50million to 3x the current bitcoin network difficulty(defense), we are talking an "evildoer" bringing 200%-300% of network power online with the sole purpose of destroying the network. There will be no recourse - for the "good guys" to match those hardware numbers it will take months to produce as noone has 20K-30K TH of hardwares just sitting there. There is also no regulation against it, i am sure there will be lawsuits and some reputational risk but all that and $50m is a drop in the bucket if an entire industry's profit is threatened. (More realistically, the banks will probably just get their lobbyist to do the work instead, see 1) ) Bitcoin can be a victim of its own success, it needs to continue the slow steady climb not become too popular too fast until its network can be secured - in my view it needs at least another 100X increase in network difficulty/power to have minimum security against the type of attack described in 2). The 2-3TH mining hardware coming Q1 2014 will be a good start. I tried to explain on the bitcoin forum you have a 12 billion market cap commodity secured by <$50 million (in network power/defense). It's insanity, but the supporters just dont get it. It's kind like carrying bricks of gold in paper bags.
The risk you mentioned are outside risks, meaning it depends on other factors and people not bitcoin itself. The 51% attack is an inside risk, meaning the protocol is faulty. I am not going to go into details (read the thread provided) but much less than 51% is enough to mess with the system. Also, two 30-30% mining pools could collide and form a bigger than 50% hash share. Not to mention you don't know who is in real control of those pools. Just for the sake of argument, if a guy holds a gun to the pool runner's head, do you think his best interest is to protect BTC? Now some other altcoins actually addressed this issue.There are some coins being actively developed to avoid this type of problem, particularly using Proof-of-Stake mining. Two of note are Peercoin and NXT. The bottomline is not to trust the masses but make a massive and foolproof protocol... TL;DR: When a system has to rely on people's goodwill, rather than correct standards and a safe algorithm, then that system is fucked... -------------------------- Here is a quote what I agree with: "Maybe a lot of you need to realize Bitcoin isn't some flawless creation that can never fail. The fact that people need to repeatedly post warnings on an online forum in order to prevent Bitcoin from being undermined/destroyed by large pools is a testament to the fact that there is something fundamentally wrong with Bitcoin that needs to be fixed. We didn't get to 42% because all of those individual miners conspired to get to that number. We arrived at this dilemma because this is the natural result of a free market and people acting on behalf of their own self interest. Sometimes people don't see the "big picture" and even when they do, they'll expect someone else to fix it, and not themselves -- no, definitely not themselves because they're too busy mining at 0% fees. This 51% will rear its ugly head over and over until either it's fixed or Bitcoin dies a horrible death. Expecting people to self-regulate is called procrastination and is futile in the long-term. Because apparently, posting these threads hasn't done anything of use. We were at 42% 2 days ago, and we're back up to 41% today. When the hype dies down, everyone switches back to GHash.IO."
ghash said that at 41% they stop accepting new miners into their pools. Other bitcoiners will move to other pools when it gets too high too. Look at them now...they are at 34% and falling. https://blockchain.info/pools?timespan=24hrs
Although I agree a 51% attack by an outside enemy sounds scary, it's not that damaging.... for the following reasons: a) its costs are enormous, compared to what the attacker can earn. b) it will be patched/solved quite fast (there are already recover-scenarios, readily coded to deploy in case that's needed). c) Many times before 51% attacks have been performed (on alt-coins) and lessons are learned how to recover quickly without much damage. If I make a quick calculation, you need about $90million currently for a 10 minute problem (understand that a 51%-attack doesn't harm the 5 year blockchain at all, so all original Bitcoin balances and transactions will stay in original state). All what is harmed is 10 minutes of transactions of the current block, maybe even a second block, before all bells go off (then point (b) and (c) will be preformed to solve and restart). In other words: why would anyone "invest" $90million currently (which value will be doubled each next 3 months!!) to disrupt 10 to 20 minutes of transactions. It will not die Bitcoin, it will be restarted with a patch (and with all original balances from before the attack). I really cannot imagine that some evil entity wants to throw away about $100million (currently). But sure: I agree for 100% that it will damage public's trust, and it will dump prices for some time, and it would be much better if the protocol could solve this in the core (but that's as difficult as solving the Byzantine Generals problem in a better way....something for a Nobel price).
And again, my point still stands: Bitcoin depends more on people's goodwill, then the security of its protocol.... Now who is to say that a bigger entity couldn't pay off the miners and take over? I am quoting from reddit: "China and/or USA could PAY cex.io $1 billion or more to DO something bad with their 51% power... essentially crash/kill the bitcoin network. Cex.io would do it for the money and it would be totally worth it for USA to do this in order to save their precious dollar."
I know at least one entity who has that kind of money. Yes, you guessed it right, the government of the USA. The reasons are obvious why they would do it. Now another thing what the attacker can do is to withhold other miners blocks, so it isn't just a 10-20 minutes problem... The question is: if some altcoins already exist that solved this problem, why don't we just use those altcoins instead? Let's switch to the more developed and more secure digital currency....And this is the beauty of new technologies, the more advanced should keep going on and the less secure should be forgotten....
I don't think we are talking the same thing. What I am saying is the network is at around 12k TH right now, you can easily mass produce 2TH miners at $1000 a pop, so an entity can easily drop 36k TH onto the network at once for $40million including data center and infrastructure cost, add on another $10million R&D to produce the asic chips (very easy) and they can blindside the network by controlling 75% in a single shot with the sole purpose of destroying the network. It will be a slow death as the network folds onto itself and there is nothing anyone can do about it. Bottom line is the difficulties are too low currently. As to why anyone want to destroy it, it's pretty obvious when billions of banking profits are threatened, it is a no brainer decision. Look what they did with the commodity infrastructure buildout driving trucks in circles before it all went to hell. This will be nothing, those guys can execute it in their sleep. PS pekelo you not getting it, there is nothing wrong with design, it's p2p open source network, majority wins.
Anyone who knows what ASICs are and how to produce them will know what bitcoin is and likely own bitcoin themselves. They catch wind of the government even ASKING engineers to produce a mass quantity of mining rigs and bitcoiners will know about it before the government can even open their checkbook.
We are talking the same thing, but I don't agree with your "too easy and too low calculations". 1) It's proves to be much harder to produce very high tech 20nm hardware in a very short timespan (especially if this should be under the radar). There are only a few manufactories in the world, fully booked for the whole year. It will cost much more than your figure. 2) The network itself grows with about 25% per 10 days. So it will be exponentially hard to create such an huge datacenter in time. As well as this datacenter should grow as well with 25% per 10 days to sustain the attack. Let alone the power and heat issues. 3) As I wrote already: the Bitcoin-network will not die without a struggle. It's programmable and will quickly be patched to defend against the evil-entity. Yes, we together can see "obvious reasons" as we both know and understand what Bitcoin will bring us. But that's NOT the case on Wallstreet at all... they still ignore and laugh at us....it will take years before they understand what hit them, and by then there's no way they could perform a 51% attack anymore, by just a few $100million.